AnthropicClaude
Claude Fable 5.2
Anthropic · Claude. Released Nov 9, 2026.
Claude Fable 5.2 is a model from Anthropic in the Claude family, released Nov 9, 2026. evals.report tracks 8 reported Claude Fable 5.2 benchmark scores across ARC-AGI-2, ARC-AGI-1, FrontierCode 2.0, AA-Omniscience v1.1: Knowledge and Hallucination Benchmark, AA-Omniscience: Knowledge and Hallucination Benchmark, Humanity's Last Exam, SWE-bench Pro, Remote Labor Index — each shown with its benchmark, metric, source status, and date, and never combined into a single ranking.
8 results
Benchmark results 8
Compare this model| Benchmark | Category | Score | Metric | Status | Date | |
|---|---|---|---|---|---|---|
| ARC-AGI-2 | Reasoning | 93.75% | accuracy | Official | Nov 9, 2026 | Details |
| ARC-AGI-1 | Reasoning | 98.5% | accuracy | Official | Nov 9, 2026 | Details |
| FrontierCode 2.0 | Coding | 50.2% | weighted score (Main) | Official | Nov 9, 2026 | Details |
| AA-Omniscience v1.1: Knowledge and Hallucination Benchmark | Reasoning | 67 | AA-Omniscience Net Score | Official | Nov 9, 2026 | Details |
| AA-Omniscience: Knowledge and Hallucination Benchmark | Reasoning | 58 | AA-Omniscience Index | Official | Nov 9, 2026 | Details |
| Humanity's Last Exam | Reasoning | 73.2% | accuracy | Verified | Nov 9, 2026 | Details |
| SWE-bench Pro | Coding | 93.0% | % resolved | Unverified | Nov 9, 2026 | Details |
| Remote Labor Index | Agents | 41.25% | automation rate | Official | Nov 9, 2026 | Details |